Journal of Natural Products Vol. 5 0 , N O . 4 , pp. 636-641, Jui-Aug 1987
63 6
STEROIDAL SAPOGENINS FROM SOLANUM SCORPlOlDE U M ALFREDON. USUBILLAGA~ and GINAMECCIA lnjtituto de InveJtigacion Quimica, Faruitad de Farmacia, Uniwrsidad de Los Anah, Merida, Venezueia ABSTRACT.-TWO
new ketosapogenins, (22R,25R)-ba-hydroxy-5a-spirostan-3-0ne
[la) and its 25s epimer [lb] were isolated from a CHCI, extract of the juice obtained from the green berries of Solanum srorpioideum. Chlorogenin E2a) and neochlorogenin [2b] were also isolated.
Solanum scorpioideum Rusby (Solanaceae) is a small tree native to the Columbian Andes. Perez Medina et al. ( l ) , in a screening of 60 Columbian Solanaceae, reported a negative test for steroidal glycoalkaloids in the fruits of this plant. The juice obtained from the green berries was extracted with CHCI,. Column chromatography of this extract yielded l a , m p 208-211°, [ a ] ~ ' ~ . ~ - 2 9 (c . 6 1.98, CHCI,) l b , m p 215-219', [a]D2'.5-52.7 (c 4.21, CHCI,) and small amounts of chlorogenin [2a) and neochlorogenin [ab]. Mass spectra and elementary analyses established the molecular formula of both l a and I b as C2,H4*04 (M+ 430). The fragmentation patterns of both substances showed prominent p e a k s at mlz 139 and 115 indicative of a spiroketal moiety (2). The ir spectra showed prominent hydroxyl and carbonyl bands, as well as the characteristic bands in the fingerprint zone attributed to the spiroketal moiety. An analysis of the ir spectra indicated that l a was a 25R spirostan because the 900 cm-' band was more intense than the 920 cm-' band (3,4); their intensities were reversed in l b indicating a 25s configuration. The I3C-nmr spectra confirmed these conclusions. The F-ring carbons of l a exhibited resonances at 109.1 pprn (C-22), 31.46 (C-23), 28.8 (C-24), 30.25 (C-25), 66.8 (C-26), and 16.98 (C-27). Those oflbshowedpeaksat 109.6ppm(C-22), 26.08 (C-23), 25.8 (C-24), 27.3 (C-25), 65.2 (C-26), and 16.02 (C-27). These assignments, obtained with help of APT spectra to discriminate carbon types, agree with those of Tori et al. (5). With the exception of these carbons all the other signals in the 13C nmr spectra of l a and l b were practically identical. The carbonyls appeared at 2 10.9 and the carbons carrying the hydroxyls at 69.7 ppm. A deuterated derivative (6) showed that the ketone functionality was flanked by methylenes (37.6 and 39.8 pprn). It was, therefore, inferred that the carbonyl was at C-3. This proposition was confirmed by the mass spectra of 3a and 3b, the ketal derivatives of l a and I b , respectively, (M+474, C2,H4,05); both substances displayed a base peak at mlz 99. It has been demonstrated (7) that steroids with an ethylene ketal at C-3 and no other substituent at ring A produce the most abundant fragment at mlz 99. If there is no substituent on ring B, the second most abundant fragment is usually at m/z 125. The absence of such a fragment in the mass spectra of 3a and 3b indicated that the hydroxyl group could be located at C-6 or C-7 in both substances. The location of the hydroxyl group was also suggested by the chemical shifts of carbons C-5 and C-7 (53.1 and 41.8 ppm, respectively). Upon acetylation of l a and I b , these signals appeared at 50.1 and 37.6 pprn in both acetates [lc and Id]. To confirm this, l a was treated with chromic acid yielding a diketone with mp 233-236", which was in good agreement with the value reported by Marker et al. for chlorogenone ( 8 ) . Elution of the column with EtOAc yielded a mixture of 2a and 2b (322 mg). The mixture was chromatographed over a small column of Si gel, but complete separation was only obtained by preparative tlc. The plates were developed 10 times with C,H,EtOAc (1:1) yielding 28 mg of 2a and 63 mg of 2b. The first of these substances, m p 268-27 lo, was identihed as chlorogenin by comparison of their physical properties with
Jul-Aug 19871
Usubillaga and Meccia:
u
0
H i
0 R3
la Ib IC Id
Rl
R2
R,
CH, H CH, H
H CH, H CH,
H H Ac Ac
R2
R3 H H Ac
Rl 2a CH, 2b H 2c CH,
(Oj nu vn
637
Steroidal Sapogenins
H CH, H
R4 H H Ac
R, R2 3a CH3 H 3b H CH,
the experimental values reported by Wall et al. (9). The melting point and ir spectrum of 2c, the diacetate of 2a,were also in fair agreement with those reported in the literature (4). The catalytic hydrogenation of la in EtOH rendered a product identical to 2a (mmp, tlc, 'H nmr, and 13C nmr). It was, therefore, concluded that the hydroxyl group in la was equatorial (&-OH). Consequently, it was inferred that Ib also had a 6 a hydroxyl. A comparison of the ir and 13C-nmr spectra of 2a and 2b permitted the inference that 2b,mp 262", was the 25s epimer of 2a,and it was, therefore, identified as neochlorogenin. The possibility that Ib would be an artifact is not likely. The epimers la and lb were obtained from the juice in about equal proportions, which is different from the equilibrium concentrations attainable by acid catalysis (10,ll).To test this assertion
Wol . 50. No . 4
Journal of Natural Products
638
and at the same time to establish the relationship between both keto-hydroxy sapogenins. pure l b was refluxed with HCl in EtOH solution . After 48 h most of the 25s compound l b had been converted to the more stable equatorial epimer [laI.. The trivial names scorpiogenin and neoscorpiogenin are proposed for these compounds . Inasmuch as the juice was stored overnight at 6" and shaken with CHC1, the following morning. it is possible that plant enzymes could have hydrolyzed the saponins originally present . Nevertheless. the Occurrence of free sapogenins is not unknown. examples having been reported by Srivastava (12). In the Solanaceae a sugar-free 3-hydroxy4-keto sapogenin has been reported by Gonzalez etal . (13). and a 3-keto sapogenin has been isolated from Solanum tovarensis (14). TABLE1. Carbon-13Chemical Shift's of (25R)-6-Hydroxy-5-spirostan-3-one[la]; (25S)-6-Hydroxy5-spirostan-3-one[lb]; (25R)-6-Acetoxy-5-spirostan-3-one[lc]; (25S)-6-Acetoxy-5-spirostan-3-one [ld]; Chlorogenine [2a]; Neochlorogenine [2bI;(25R)-6-Hydroxy-5-spirostan-3-ethylenedioxy [B]; and (25S)-6-Hydroxy-5-spirostan-3-ethylenedioxy [3b]
Compounds
Carbonatom
c-1
. . . . . . .
c-2 . . . . . . .
c-3 . c-4 . c-5 . C-6 . c-7 . c-8 . c-9 . c-10 . c-11 . c-12 . C-13 . C-14 . C-15 . C-16 . C-17 . c-18 . C- 19 . c-20 . c-21 . c-22 . C-23 . C-24 . C-25 . C-26 . C-27 .
. . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
. . . . . . . . .
la
Ib
IC'
38.6 37.6 210.9 39.5 53.1 69.7 41.8 34.0 53.5 36.5
38.5 37.6 213.9 39.5 53.1 69.7 41.8 34.0 53.5 36.5
38.3 37.5 2 10.6 39.4 50.1 72.2 37.6 33.8 53.4 36.8
38.6 37.6 210.9 39.5 50.1 72.2 37.6 33.8 53.5 36.8
21.1
21.1
21.1
21.1
39.8 40.6 55.9 31.8 80.6 62.3 16.0
39.8 40.6 55.9 31.8 80.7 62.3 16.0 12.7 42.3 14.1 109.6 26.1 25.8 27.1 65.1 16.3
39.7 40.6 55.9 31.7 80.5 62.4 16.2 12.6 41.7 14.3 109.0 31.5 28.8 30.3 66.8 16.9
39.8 40.6 55.9 31.7 80.6 62.3 16.0 12.3 41.7 14.1 109.6 26.1 25.8 27.3 65.2 16.3
. . . . 12.7 . 41.8 . 14.3 . 109.1 . 31.5 . 28.8 . . . . . . 30.3 . . . . . . 66.8 . . . . . . 16.9
2a 37.3 29.6 71.7 30.8 51.8 69.3 41.6 33.9 53.9 36.3 21.0 39.9 40.6 56.1 31.6 80.8 62.2 15.9 13.4 42.1 14.2 109.2
31.5 28.8 30.2 66.8 16.9
2b
3a
3b
37.3 29.5 71.7 30.8 51.7 69.2 41.6 34.0 53.9 36.3 21.0 39.8 40.6 56.1 31.6 80.8 62.2 15.9 13.3 42.2 14.0 109.7 26.0 25.8
36.2 31.2 109.2 32.2 50.8 69.3 42.2 34.0 53.8 36.4 21.0 39.9 40.6 56.0 31.7 80.6 62.2 16.0 12.5 42.0 14.2 109.0 31.5 28.8 30.3 66.8 17.0
36.3 31.1 109.1 32.2 50.8 69.4 42.1 34.0 53.7 36.3 21.0 39.9 40.6 56.1 31.8 80.5 62.3 16.0
27.1
65.2 16.2
12.5
42.3 14.1 109.1 26.1 25.8 27.3 65.1 16.3
'Acetates: IC.20.8(Me).170.3(carbonyl). Id. 20.9(Me). 170.4(carbonyl).
This study was undertaken based on the report of Perez Medina et a1. (1) that no glycoalkaloids were present in the fruits of S . scorpioideum . Such anomaly is sometimes an indication that unusual steroidal alkaloids might be present (15.16). but in this case no alkaloids were found . EXPERIMENTAL
GENERL-TIC was performed on Si gel G plates;the spots were visualized by sprayingwith dilute
Jul-Aug 19871
Usubillaga and Meccia:
Steroidal Sapogenins
639
H,S04 and charring at 110". Melting points were measured on a Fisher-Johns hot-stage and are uncorrected. Optical rotations were measured on a Zeiss polarimeter Model 53 187 using a sodium lamp. The 'H-nmr and I3C-nmr spectra were determined in CDCI, with TMS as the internal standard in a Varian spectrometer Model FT-80A. Chemical shifts are expressed in 6 values ('H) and ppm (I3C-nmr). In addition to completely decoupled spectra, APT spectra were obtained for carbon type discrimination. The ir spectra were recorded on a Perkin-Elmer Model 377 spectrometer as KBr discs. The mass spectra were performed in a Jeol Model JMSG-2 apparatus at 70 ev using direct inlet. Microanalyses were made at Simon Bolivar University, Caracas. PLANT MATERIAL, EXTRACTION, AND CHROMATOGRAPHY.--Green berries (7.8 kg) Of s. JCOYpioiahm were collected in August, 1985, in Usaqukn, a suburb of Bogota. A voucher specimen is kept at the MERF Herbarium (Herbarium of the Faculty ofPharmacy, ULA, Merida, Venezuela). The same day of collection the fruits were coarsely ground, and the juice, obtained by pressing the pulp, was shaken with CHCI,. Evaporation of the solvent at reduced pressure yielded 17.2 g of extract which was treated over a Si gel column (500 g). Elution was started with C6H6, followed by C6H&tOAc mixtures; 500-ml fractions were taken and examined by tlc. From fraction 26 (C6H6-EtOAc, 10: 1) 232 mg of l a was obtained as crystals, mp 204-208". Recrystalization from MeOH afforded crystals with mp 208-21 lo, pure on tlc, [ a ) ~ ~ ' - ~ - 2 9(c ,1.98, 6 CHCI,); ir v max 3420, 1690, 985, 920, 900 cm-'; ms m/z 430 (M+, 21.6% C27H4204),394 ( l l % ) , 358 (19.3%), 316 (34%), 298 (23%), 269 (38%), 139 (base peak, C,H,,O), 115 (47%); 'H nrnr (80 MHz, CHCI,) 6 0.78 (3H, s, 18-Me), 1.02 (3H, s, 19-Me), 0.94 (3H, d, J=7 Hz, 2l-Me), 0.75 (3H, d, J=7 Hz, 27-Me), 2,72(1H,dd,J=16and4Hz), 3.35(1H,m,C16a-H), 3.40(1H,s,OH),4.35(1H,m, C6-HOH); '3CnmrseeTable 1. FoundC75.52%, H9.61%; C27H4204requiresC75.35%, H9.77%. Fractions 37/45 yielded 190 mg of l b mp 215-219', [ a ] ~ " . ~ - 5 2 . 7 (c 4.21, CHCI,); ir v max 3420, 1690, 990, 922, 899 cm-'; ms m/z 430 (M+, 16%, C27H4204)r348 (14%), 316 (21%), 301 (lo%), 298 (8%), 288 (22%), 139 (base peak, C9HI5O) 115 (43%); 'H nrnr (80 MHz, CHCI,) 6 0.76 (3H, s, 18-Me), 1.02 (3H, s, 19-Me), 1.05 (3H, d, J=7 Hz, 21-Me), 0.96 (3H, d, J=7 Hz, 27-Me), 2.70 ( l H , dd, J= 16 and 4 Hz), 3.32 ( l H , t, J =12 Hz, C16-aH), 3.40 ( l H , s, OH), 3.93 ( l H , dd, J= 10 and 2 Hz),4.35 ( l H , m, C6-HOH); 13Cnrnr see Table 1; Found: C 75,46%, H 9.62%; C27H4204 requires: C 75.35%, H 9.77%. From fractions 27/36 594 mg o f l a + l b was obtained. This mixture was later separated by thick layer chromatography on 1 mm thick 2 0 x 4 0 cm plates. Five developments with C,&-EtOAC (2: 1) were employed to obtain 2 10 mg of l a and 255 mg of I b . Fractions 60/80 were pooled, and the residue (322 mg) was treated over a small Si gel column (50 g). Elution with C6H6-EtOAc (5: 1) yielded a mixture of 2a and 2b (127 mg). Complete separation of both substances was obtained by preparative tlc using 0.5 mm 20 X 20 cm plates. Ten developments with C&EtOAc (1: 1) yielded 28 mg of 2a and 63 mg of 2 b . The first of these substances [2a] crystallized from MeOH as fine needles, mp 268-271", [ a ] ~ ~ l . ~ - 4 (r2 "0.40, MeOH); ir v max 3400, 985, 922, 900 cm-'; 'H nmr (80 MHz, CHCI,) 60.77 (3H, s, 18-Me), 0.84 (3H, s, 19-Me), 0.94 (3H, d, J=7 Hz, 21-
Me),0.78(3H,d,J=7Hz,27-Me),2.15(1H,s,C3-~OH),3.35(1H,m,C16-aH),3,40(1H,s aOH), 4.35 ( l H , m, C6-HOH); '3CnmrseeTable 1; Found:C75.24%, H 10.31%;C27H4404requires: C 74.96%, H 10.25%. This compound was identified as chlorogenin. Its mp and optical rotation agree with those reported in the literature (9). The second substance [Zb], crystallized from (CH,),CO mp 262', [a)D2'.5-610(r 3.2, MeOH); ir v max3400, 1055,985,920, and900cm-'; 'Hnrnr(80MHz,CDCI,)80.76(3H, s, 18-Me),0.85(3H, s, 19-Me), 1.07 (3H, d, J=7 Hz, 2l-Me), 0.97 (3H, d, J=7 Hz, 27-Me), 2.15 ( l H , s, 3B-OH), 3.35 ( l H , t, J= 10 Hz, C16-aH), 3.73 ( l H , s, 6a-OH), 3.92 ( I H , dd, J= 11 and 4 Hz), 4.35 ( l H , m, C6HOH); 13Cnrnr see Table 1; Found: C 75.40%, H 10.12%; C27H4404requires: C 74.9696, H 10.25%. A comparison of the I3C-nmr spectrum of this substance with the spectrum of 2apermitted identification as neochlorogenine. ACETYLATION OF la.-Anhydrous pyridine and Ac,O were added to 60 mg of l a a n d left overnight at room temperature. The following morning, iced H,O was added, and the precipitate was filtered and washed with H,O. The acetate ICcrystallized from Me,CO as colorless needles, mp 164-165'; ir v max 1740, 1708, 1250 cm-'; 'H nrnr (80 MHz, CDCI,) 6 0.75 (3H, s, 18-Me), 1.04 (3H, s, 19-Me), 0.92 (3H, d, J=7 Hz, 27-Me), 1.08 (3H, d, J = 7 Hz, 21-Me), 1.99 (3H, s, acetate); I3C nmr see Table 1. ACETYLATION OF l b . - h the same manner as above, 60 mg of l b was acetylated. The acetylated derivative ldcrystallizedfromMe,CO, mp 193-195"; irvmax 1735, 1705, 1245,985,920,900cm-'; 'H nmr (80 MHz, CDCI,) 6 0.80 (3H, s, 18-Me), 1.10 (3H, s, 19-Me), 0.98 (3H, d, J=7 Hz, 27-Me), 1.08 (3H, d, J=7 Hz, 2l-Me), 2.00 (3H, s, 0-Acetate); I3C nmr see Table 1. DEUTERATION OF 1b.-Tetrabutylammonium bromide (50 mg) and l b (80 mg) were dissolved in 10 ml of freshly distilled dry CH,CI,. This mixture was shaken during 3 days at room temperature with 5
640
Journal of Natural Products
Wol. 50, No. 4
ml of D,O containing 5% NaOD under dry N, atmosphere. The organic layer was shaken twice again with NaOD solution during 24 h. Finally the organic layer was dried over Na,S04 and evaporated. The residue was purified over a small s i gel column that was eluted first with C&. Elution with a mixture of C6H6EtOAc (1: 1) afforded 45 mg ofpure (25S)-6a-hydro~y[2,2,4,4-~H~]-5a-spirostan-3-one, mp 209-2 11'; ir Y max 3410, 1675 cm-'; 'H nmr(80 MHz, CDCI,) 6 0.75 (3H, s, 18-Me), 0.99 (3H, s, 19-Me), 1.07 (3H, d, J=7 Hz, 2 1-Me), 0.93 (3H, d, J=7 Hz, 27-Me), 3.32 ( l H , t , J = 12 Hz), 3.92 ( l H , dd, J= 12 and4 Hz), 4.35 ( l H , m); I3C nmrC-l(38.4), C-2 (missing), C-3 (210.6), C-4(missing), C-5(53.0), C-6 (69.7), C-7 (41.8), C-8 (34.0), C-9 (53.6), C-10 (36.5), C-11 (21.1), C-12 (39.8), C-13 (40.6), C-14 (55.9), C-l5(31.8),C-16(80.7),C-17(62.3), C-18(16.3),C-19(12.7),C-20(41.8),C-21(14.1),C-22 (l09.6), C-23 (26. l), C-24 (25.8), C-25 (27. l), C-26 (65.2), C-27 (16.0). (25S~3-ETHYLENEDIOXY-5a-SPIROSTAN-Q-OL [3b].-To obtain this ketal derivative the procedure of Herzog et uf. (17) was followed. 50 mg of l b was dissolved in 25 ml ofdry C6H6. To this solution 20 mg of P-TsOH and 0.2 ml of ethylene glycol were added, and 15 ml of C6H6 was distilled to eliminate H,O by azeotropic distillation. The organic layer was washed with a solution of Na,CO3, dried over anhydrous Na,S04, and distilled to dryness. Crystallization from MeOH yielded 3b as fine needles, mp 253-260". ir no carbonyl absorption; ms m/z 474 (M+, 18%, C29H4605),456 (3%), 402 (8%), 360 (35%), 33 1 (6%), 268 (8%), 25 1 (7%), 139 (64%),99 (basepeak, C,H,O,); 'H nmr(80 MHz, CDCI,) 6 0.75 (3H, s, 18-Me), 0.80 (3H, s, 19-Me), 1.07 (3H, d, J=7 Hz, 2l-Me), 0.92 (3H, d, J=7 Hz, 27Me), 3.9 (4H, m); I3C nmr see Table 1.
(ZSR)-3-ETHYLENEDIOXI-5a-SPIROSTAN-6a-OL [%].-A solution of 50 mg of l a in 25 d of dry C6H6 was treated in the same manner as explained above to obtain its ketal derivative mp 24 1-245', ir no carbonyl absorption; ms d z 474 (M+, 25%, C29H4605),456 (7%), 402 (10%). 360 (37%), 331 (8%), 269(15%), 139(51%), 99(100%); 'Hnmr(80MHz,CDC13)60.75(3H, s, 18-Me), 0.83 (3H, s, 19-Me), 0.94 (3H, d, J=7 Hz, 21-Me), 0.79 (3H, d, J=7 Hz, 27-Me), 3.90 (4H, m); I3C nmr see Table 1.
[a],
CHROMICACID OXIDATION OF la.-Kiliani's reagent was added drop by drop to a solution of 220 mg of l a in Me,CO. After 30 min at room temperature, the reaction was complete, and a few drops of MeOH were added to destroy excess reagent. After the usual workup, the product crystallized from Me,CO and its mp (233-236") was found to be practically identical to the mp reported for chlorogenone (8); irvmax 1720and 1700cm-'(twocarbonyls); 'Hnmr(80MHz. CDC13)60.96(3H,s, 18-Me), 1.02 (3H,s, 19-Me), 1.07(3H,d,J=7Hz,2l-Me),0.85(3H,d,J=7Hz,27-Me),3.55(1H,dd,J=24and
lOHz),4.2(lH,d,J=6Hz),4.75(2H,m). CATALYTICHYDROGENATION OF la.-A solution of l a (50 mg) in EtOH (25 ml) was mixed with 50 mg ofprereduced PtO,, suspended in EtOH, and shaken during 4 h under an atm ofH, at 50 psig. The catalyst was filtered off, and the solution was evaporated to dryness. The residue crystallized from MeOH, mp 268-270', was identical to 2a (mmp, tlc, 'H nmr, I3C nrnr). ISOMERIZATION OF 1b.-A solution of l b (20 mg) in EtOH (4.5 ml) was mixed with 0.75 ml of conc. HCI and refluxed for 48 h. The reaction was followed by tlc, and it was stopped when no further change in the relative concentrations of 25s (Rf=0.27) and 25R (Rf=0.32) isomers was observed. The reaction mixture was taken to dryness and applied to a 0.5 mm thick Si gel plate (20 X 20 cm); the plate was developed three times with C6H6-EtOAc (1: 1). The bands were made visible spraying H,O and extracted with CHC1,IMeOH. The upper layer yielded 9 mg of a compound identical to l a (tlc, mp, and ir); from the lower layer a very small amount (0.8 mg) ofthe original substance [lb]was recovered. Several by-products were visible on the tlc plate, but they were not investigated. ACKNOWLEDGMENTS Financial support from Consejo de Desarrollo Cientifico, Humanistic0 y Tecnologico (CDCHT-
ULA)is gratefully acknowledged. The authers also thank Dr. Juan Martinez and Dr. Enrique Cucas of Univeaidad Nacional de Colombia (Bogota) for their kind help in the collection and extraction of the plant material. They are also indebted to Dr. Edgar Moreno of Federacidn National de Cafeteros (Colombia), for the mass spectra, to Dr. Antonio Zapata of Univeaidad Simon Bolivar (Caracas)for microanalysis, and to Dr. Miguel Alonso, IVIC (Caracas), for optical rotation measurements. LITERATURE CITED 1. L.A. Perez-Medina, E. Travecedo, and J.E. Devia, Pluntu Mal., 1 2 , 478 (1964). 2. H . Budzikiewicz, C. Djerassi, and D.H. Williams, "Structure Ellucidation ofNatural Products by Mass Spectrometry," Vol. 2, Holden Day, San Francisco, 1964, p. 110.
Jul-Aug 19871 3. 4.
5. 6. 7.
8. 9. 10. 11. 12. 13. 14. 15. 16. 17.
Usubillaga and Meccia:
Steroidal Sapogenins
64 1
M.E. Wall, C. Roland Eddy, M.L. M d l e n a n n , and M.E. Klumpp, Anal. Chem., 2 4 , 1337 (1952). C . Roland Eddy, M.E. Wall, and M. Klumpp Scott, Anal. Chem., 2 5 , 266 (1953). K. Tori, S. Seo, Y. Terui, J. Nishikawa, and F. Yasuda, Tetrabedron k t t . , 2 2 , 2405 (1981). C. Starks,]. Am. Chem. Sor., 9 3 , 195 (1971). G . von Mutzenbecker, Z. Pelah, D.H. Williams, H . Budzikiewicz, and C. Djerassi, Steroid, 2 , 475 (1963). R.E. Marker, E.M. Jones,and D.L. Turner,]. Am. Chem. Sor., 6 2 , 2537 (1940). M.E. Wall, M.M. b i d e r , E.S. Rothman, and C. Roland Eddy,]. Biof. Chon., 198, 533 (1952). R.E. Marker, A.C. Shabica, and D.L. Turner,]. Am. Chem. Sor., 6 3 , 2274 (194 1). M.E. Wall, S. Serota, and L.P. Witnauer,]. Am. Chem. Sor., 77, 3086 (1955). S.K. Srivastava and S.D. Srivastava, Phytorhemistry, 1 8 , 1758, (1979). A.G. Gonzalez, C.G. Francisco, R. Freire, R. Hernandez, J.A. Salazar, E. Suarez, A. Morales, and A. Usubillaga, Phytorhmistry, 1 4 , 2483 (1975). A. Morales Mendez, C. Riera, andL. Moreno, Revistadela Facultadde F a m i a (ULA-Merida),1 5 , 133 (1974). A. Usubillaga, C. Seelkopf, J. Katle, J. Dale, and B. Witkop,]. Am. Chem. Sor., 9 2 , 700 (1970). A. Usubillaga, A. Paredes, P. Martinod, and J. Hidalgo, Planta Med., 2 3 , 286 (1973). H.L. Herzog, M.A. Jevnik, P.L. Perlman, A. Nobile, and E.B. Hershberg,]. Am. Chem. SM.,7 5 , 266 (1953).
Received 3 Novermber I986
